Alternative Law Journal
GAVIN TURBETT[*] explains some aspects of DNA profiling.
Any discussion of the use of forensic DNA profiling necessarily involves the consideration of some complex scientific questions. The aim of this article is to briefly address, from a forensic scientist’s point of view, some of the matters raised by Webb and Tranter and Kellie in their articles published in this edition. In so doing I shall attempt to correct some commonly held misconceptions about what a DNA sample is and, more importantly, what it is not.
The present method of forensic DNA profiling does not enable the analyst to determine anything more about the subject of the sample than whether the donor is male or female, and the statistical probability of that sample matching another sample taken. In order to understand this, it is necessary to explain some facts about DNA.
DNA, or Deoxyribonucleic Acid, may be thought of as being composed of ‘coding’ and ‘non-coding’ regions. The coding regions are known as genes. Genes are the sequences of DNA that allow cells to create a protein product. Estimates of the total number of genes in the human genome range from as few as 30,000 to more than 90,000, with approximately 26,000 documented so far. Non-coding DNA has no known function, and accounts for approximately 97% of the total human genome. There is very little room for changes to occur safely in coding DNA. In contrast, mutations may occur in non-coding DNA without affecting the host and as a consequence it can vary significantly between individuals.
One class of non-coding DNA consists of repetitive sequences. The repetitive sequences now widely used for forensic purposes are known as Short Tandem Repeat (STR) sequences. It is the number of tandem repeats that varies between individuals. The number of repeats affects the sizes of the DNA fragments. It is this variation in the STR sequences between people and the relative ease by which the size variation can be measured that makes them so useful in forensic DNA profiling. A good overview of STRs used for human identity testing is the Short Tandem Repeat DNA Internet DataBase. There are literally tens of thousands of STRs scattered throughout an individual’s genome.
The system most commonly used for forensic DNA profiling in Australia is called Profiler Plus, which analyses nine STRs at once. In total, the Profiler Plus kit analyses only approximately 1 part in 1.5 million of our genome. Similar DNA profiling kits are used in other parts of the world and many analyse the same STR markers. Legal rulings concerning the admissibility of this technique for forensic DNA profiling are available online. Because Profiler Plus is used across Australia, results generated in one lab are directly comparable to the results generated in another. This of course, is an absolute necessity for participation in the CrimTrac National Criminal Investigation DNA Database (NCIDD).
STR sequences are analysed using a method called the Polymerase Chain Reaction (PCR). PCR may be thought of as a method for ‘photocopying’ DNA, so that initially small amounts of target sample can be copied until there is sufficient DNA for analysis. Explanations of the PCR process may be found online.
A forensic DNA profile has no known prognostic, diagnostic or medical value. It cannot be used to predict whether or not an individual is susceptible to certain diseases (such as Alzheimer’s or specific cancers), or is predisposed to criminal behaviour, alcoholism or any other trait.
Apart from the nine STR loci, Profiler Plus also examines the Amelogenin locus, which is found on the both of the sex (X and Y) chromosomes. This allows the determination of the individual’s sex. This is the only physical characteristic that may be determined by Profiler Plus.
Accordingly, concerns raised about discrimination must be tempered by an understanding of exactly what regions of the genome a forensic DNA profile is testing. Further, the distinction between coding and non-coding DNA must be borne in mind when discussing potential concerns about the ‘central dogma’.
Webb and Tranter refer to a ‘series of assumptions’ about DNA known as the central dogma. This is an oversimplification. The central dogma of molecular biology has been arrived at over decades of research. It is that:
• DNA can be duplicated to create identical copies of itself. This duplication is required before each round of cell division, so that each daughter cell receives a full copy of DNA.
• the genes (coding DNA) are transcribed into an intermediate molecule known as ribonucleic acid (RNA).
• RNA is then translated into protein products. RNA is read in sets of three bases known as codons. The sequence of each codon dictates the order of the amino acids used to build the protein.
Webb and Tranter suggest that the central dogma is struggling to retain credibility ‘in the face of contrary evidence’. This may be arguable, but as only non-coding DNA is used for the purposes of forensic testing, their concerns about the interrelationship of the central dogma (with its focus on coding DNA) and the use of forensic testing in the criminal justice system are, with respect, groundless. Webb and Tranter also state that ‘the problem of the central dogma for lawyers is that it compresses the phenotype to the genotype’. The Profiler Plus system analyses areas of the genome that have absolutely no bearing on the phenotype. Therefore, there is no reason why phenotype and genotype should or could be related to each other. Likewise, the present method of DNA testing is in no way related to the possible search for a ‘criminal gene’.
There are no so called ‘criminal genes’. What has been determined is that some people have variations in the sequences of specific normal genes and that these variations can affect the behaviour of the individual, most probably via affecting brain chemistry.
It has been estimated that at least 40%, and possibly far higher, of all of our genes are involved solely in the processes of building, connecting and maintaining the brain. The variation could be as subtle as a single DNA base changed in a gene that is thousands of bases long. Because genes ultimately encode a protein product, the change may result in a different amino acid (the building blocks of proteins) being inserted into the protein at a particular location. This may then affect the shape and function of that protein, which may then affect the host.
While it has been shown that variations in the level of a single gene product can be correlated to criminal activity, others argue that because human behaviour is so complex and so irrevocably bound to environmental factors, multiple genes are likely to be involved and that it may not be possible to attribute specific behavioural patterns to any one gene.
If we were to look carefully enough, every individual would be found to have occasional mutations in various genes. In most cases however, those mutations will be completely silent, in that they have no effect on function or behaviour.
In any event, even if a criminal gene could be identified, it would not be detected by current forensic testing systems such as Profiler Plus, which only tests non-coding DNA.
In my view the current system of forensic DNA profiling should not be viewed differently to any other forensic technique. A forensic DNA profile cannot provide any more information about the subject’s genetic heritage than other identification methods such as fingerprinting. Historically, it is not argued that fingerprinting is discriminatory; similarly DNA profiling does not discriminate.
Both articles raise concerns about the competency of analysts, reliability of techniques, and whether testing laboratories conform to accepted testing procedures.
Australian forensic biology laboratories have taken a great deal of care in establishing clear guidelines for the performance of forensic DNA analysis. These guidelines have been established in conjunction with the National Association of Testing Authorities, Australia (NATA), which is the sole government-endorsed accreditation body for establishing and assessing competent laboratory practice.
Forensic laboratories operate under Australian Standard ISO/ IEC 17025: 1999 General Requirements for the Competence of Testing and Calibration Laboratories as well as the Supplementary Requirements for Accreditation in the Field of Forensic Science. To achieve accreditation, the forensic laboratory must be able to demonstrate compliance with the NATA guidelines and undergoes inspection by NATA assessors. An accredited laboratory must continue to comply with all accreditation requirements and is re-evaluated every two years. NATA also requires and monitors the performance of accredited forensic science laboratories in external proficiency testing programs. Because all Australian forensic labs are participating in the same proficiency tests and are therefore analysing the same test samples, any effects of variability in testing procedures and equipment is negated.
It is true, as Webb and Tranter state, that DNA ‘may be damaged as a result of age, exposure to the elements and other substances.’ However, t his does not invalidate the science of DNA profiling any more than the science of fingerprinting is invalidated because fingerprints may become smudged. When DNA is sufficiently degraded, it starts to fail in the DNA profiling system. The larger DNA fragments fail first because longer stretches of DNA are statistically more likely to suffer a break than a small stretch. The loci chosen for Profiler Plus are all as small as possible to minimise the effects of DNA degradation. Degraded DNA will give what is known as a partial DNA profile, where a result can only be recorded at some loci. This has the effect of reducing the statistical weight that may be placed on that partial DNA profile.
The Human Genome Mapping Project announced a working draft of the entire human genome sequence in 2000. All that remains now is to fill in the gaps in the sequence. Once this has occurred, a comprehensive analysis of all human genes can begin. Many genes for physical characteristics such as hair and eye colour have already been identified and it is likely that many more will be identified in the future. There are inherent risks with this approach. While there are significant genetic components to physical characteristics such as height, weight and skin tone, these characteristics are also significantly affected by environmental factors such as nutrition and sun exposure. Furthermore, physical features such as hair and eye colour are readily disguised.
Despite this, using DNA recovered from a crime scene to build up a likely physical description of the offender may be possible in the future. This will be most relevant in cases where the offender is unknown, there are no witnesses and a matching DNA profile is not already on the National Database. Any ‘identification’ made on the basis of physical characteristics would be confirmed or refuted with conventional forensic identification methods.
At this time, a forensic DNA profile cannot be used to determine the ethnicity of the donor, although attempts are being made to achieve this. As the human genome is analysed, gene sequences that confer specific ethnic traits will be identified and could be used to determine the ethnicity of the donor. This information, along with other physical characteristics would be used to focus police investigations. Such ability carries with it its own complications, such as issues of mixed ancestry and wide variations in physical characteristics resulting from environmental factors.
It is almost impossible to predict what the future of DNA testing will bring. I agree with the authors that there is a need to be vigilant, to ensure that the technologies developed in the future are not used inappropriately, whether for forensic science, medicine or biotechnology. Science itself is neither good nor evil. It all depends on how it is used. In the case of expert witnesses in court, it may equally depend on how it is presented and how well it is understood. As Kellie states: ‘A jury is less fallible when all the evidence is before it and it is facilitated towards a full understanding of that evidence’. In order to achieve this laudable end, it is encumbent on those of us practising in the areas of forensic science and ciminal law to work together to ensure that all parties to the criminal justice system have a full understanding of forensic DNA profiling.
[*] Gavin Turbett is Scientist in Charge, Forensic Biology Laboratory, PathCentre, Perth.
©2001 Gavin Turbett
 Short Tandem Repeat DNA Internet DataBase <http://www. cstl.nist.gov/div831/strbase/> .
 Applied Biosystems web site <http://www.appliedbiosystems.com/ products/productdetail.cfm?id=100> .
 DNA Litigation Legal Support Page <http://denverda.org/legalResource/ legalresource.htm> .
 <http://www.crimtrac.gov.au/dna.htm> .
 Biology Animation Library <http://vector.cshl.org/resources/ BiologyAnimationLibrary.htm> Principle of the PCR <http://allserv. rug.ac.be/~avierstr/principles/pcr.html> .
 af Klinteberg, B., ‘Biology, norms, and personality: a developmental perspective’, (1996) . 34 Neuropsychobiology 146-54.
 McGuffin, P., Riley, B. and Plomin, R., ‘Towards Behavioural Genomics’, (2001) Science 291 (5507): 1232-49 <http://www. sciencemag.org/cgi/content/full/291/5507/1232> .
 <http://www.nata.asn.au/> .
 Human Genome Mapping Project <http://www.ornl.gov/hgmis/> .
 Shriver M.D., Smith M.W., Jin L. and others, ‘Ethnic-Affiliation Estimation by Use of Population-Specific DNA Markers’, (1997) 60 Am J Hum Genet 957-64; Lowe, A.L., Urquhart, A., Foreman, L.A. and Evett, I.W., ‘Inferring Ethnic Origin by Means of an STR Profile’, (2001) 119 Forensic Sci Int 17-22.